The presence or absence of key periodontal pathogens, either alone or in combination, may provide important prognostic information about the periodontal health of a patient or site or concern. For this determination, the total number of pathogens and the relative percent of pathogens to total bacterial load may be required for microbial data to be meaningful. To establish the relationship of the microflora to disease activity, a longitudinal study is planned. The specific aims of this proposal are to: 1. Validate the use of universal DNA probes for detecting total bacteria in a plaque specimen, as it compares to a previously validated method for measuring ATP levels in a plaque specimen. 2. Establish the predictive value of microbial data for periodontal disease by monitoring, microbiologically and clinically, sites prior to and during attachment loss. The microbial monitoring will be undertaken using previously referenced whole genomic probes to identify Bacteroides gingivalis, Bacteroides intermedius and cloned probes to Actinobacillus actinomycetemcomitans as well as the probes developed in the Phase I portion of this application to Eikenella corrodens, Wolinella recta, and Fusobacterium nucleatum. Clinical monitoring will utilize techniques of Magnusson et al. (1988) and Gibbs et al. (1988) using computer assisted fixed force periodontal probe developed by the investigators at U. of Florida.